the problem is real and underappreciated. right now, most agent frameworks handle context like this: load memory at session start. stuff it into the prompt. when the window fills, summarize and compress. hope the important parts survive. the paper calls this out directly. once
RESEARCH
-
Unix principle applied to AI context
By
–
"everything is a file" becomes "everything is context." CSIRO Data61 and ArcBlock published a paper applying the oldest Unix principle to the newest AI problem. they propose treating memory, tools, knowledge, and human input as a mounted filesystem that agents browse at runtime
-
Reasoning requires basic knowledge for effective search guidance
By
–
You still need a bit of knowledge in any reasoning, just to guide you in where to look. For instance, you'll do better search on geopolitics if you know a bit about the general trends. Qwen just demonstrated that this base level of knowledge can be attained in a very compact
-
Jinjie Ni Joins Google DeepMind as Research Scientist
By
–
Congrats and welcome @NiJinjie to the center of AGI (🇸🇬 branch). Taking highly technical and capable researchers like this and giving them a chance to be at the frontier to make tons of impact for Gemini has been one of the most rewarding things of founding & building GDM in SG. Jinjie Ni (@NiJinjie) Life update: I’ve joined @GoogleDeepMind as a research scientist to work on ✨gemini scaling and RL, under the leadership of Yi Tay (@YiTayML) and Quoc Le (@quocleix). I feel extremely fortunate to be on the critical path towards AGI and can't wait to help push the frontier of gemini capabilities! 🚀 — https://nitter.net/NiJinjie/status/2028471474265420141#m
-
4B model rivals 120B on knowledge, ideal for agentic core
By
–
It's crazy that a 4B model can match GPT-OSS-120B **even on knowledge benchs** like MMLU and GPQA! Knowledge was long a limitation of smol models ; not anymore. This 4B could be a good candidate for the "agentic core" model described by @karpathy
-
QueryWeaver GitHub Project Promotes AI-Related Categories
By
–
QueryWeaver GitHub: (don't forget to star )
-
Volume 322 AAAI Workshop on Audio-Centric AI Proceedings Now Available
By
–
Volume 322 proceedings.mlr.press/v312 Proccedings of the AAAI Workshop on Audio-Centric AI Is now available on PMLR.
-
Understanding Temperature in AI Models: Beyond Creativity vs. Determinism
By
–
Great breakdown! One thing worth adding: temperature isn't just about creativity vs. determinism. It fundamentally reshapes the probability distribution over the entire vocabulary. A practical tip: instead of jumping straight to high temperature for 'creativity,' try
-
AI Agents in Practice — Design, Implement, and Scale Autonomous AI Systems
By
–
See the latest book from @AltoValentina
, released by @PacktDataML at http://
amzn.to/4p98LYl "AI Agents in Practice — Design, Implement, and Scale Autonomous #AI Systems for Production" Book Description (from Amazon): As AI agents evolve to take on complex tasks and operate -
Practical Guide to Building AI Agents with LLMs and Knowledge Graphs
By
–
"Building AI Agents with LLMs, RAG, and Knowledge Graphs — A practical guide to autonomous and modern AI agents" See it at http://
amzn.to/4622k2h via @PacktDataML